What does it mean that the true effect size is not known? I just read this statement:

In neuroscience the true effect size is not known

What does this mean? 
In contrast, what does it mean that where the true effect size is known?
 A: Essentially, it means that the answer is not known before hand, and is arguably impossible to know. Consider, for example, the impact of exposure to a certain chemical on the development of Parkinson's Disease.
We have no basis of knowing that "Those exposed to Chemical X are twice as likely to develop Parkinson's Disease as those who are not" in the sense of knowing that that is capital-T True. The effect - in this case that the relative risk of developing Parkinson's due to exposure to X is 2.00.
We can estimate it from studies, but if the study finds the effect is 1.88, or 2.05, or -0.68, we have no way of telling if they're "right". Over time, and many well-run studies, we can become more confident we're in the neighborhood of right, but we still don't know the capital-T Truth.
In contrast, some studies, especially simulation based studies, set a fixed effect. This can be to test statistical methods (can your statistical test/software/etc. give me back the answer I know to be true?) or to ask questions about "If this effect is true, what does that mean?"
A: It means that when we begin studies, we do not know the size of the effect we are looking for. The true effect size is almost never known. We are usually trying to estimate the effect size that exists in our population through the effect size found in our sample. One way to "know" the effect size is to estimate it from literature on that effect.
